An Introduction to Microsimulation Tax-Benefit Models

Incentives and Fiscal Neutrality

As you saw in Chapter 1 of the textbook, mainstream economics starts from the premise that a properly functioning competitive market economy is the most efficient way of organising production and consumption for a modern society.

Economists often think of an economy as a “price system”: where the market works well, prices reflect the relevant costs (“at the margin”) of doing a thing, and people following these prices will be let “as if by an invisible hand”¹ to do the best thing. “At the margin” is important here - it’s the price of doing a little bit more of something that’s the key thing - the price of the next litre of petrol you buy, or the wage for the next hour you work.

In designing a tax and benefit system, one common objective is therefore to distort market incentives as little as possible. There is a presumption in favour of low marginal tax rates on goods so the prices of things don’t move away from their efficient level too much. In this context we’d consider labour to be a good that’s bought and sold at a wage, so the presumption is for low marginal tax rates on earnings. (I’ll discuss marginal versus average tax rates in more detail we have the simulation model in front of us)².

As well as low marginal rates, there is also a presumption in favour of broad based taxation. Suppose a particular brand of baked beans was taxed, but all other brands were not. Since brands of beans are likely close substitutes, even if the tax rate was low, the taxed brand of beans would likely be wiped out - a pretty severe distortion. A tax on all beans would be less distortionary, a tax on all food better still, and a tax on everything best of all from this point of view. In the UK, most food, children’s clothes and books are not subject to our 20% Value Added Tax, whilst most other goods are. Economists at the Treasury have long had their eye on these exemptions, whilst politicians and their advisors have jealously guarded them, on both distributional and electoral grounds. (Both sides are right in their own way)³.

Sometimes subtleties in the tax code matter more than the headline tax rates. Consider corporation taxes - taxes on profits of companies. If a company is operating efficiently, and making the most profit it can (maximising profits), then in principle it has no reason to change what it’s doing if a proportion of those profits are taken away - its before-tax plan is still the best it can do. Neutrality in this case means careful attention to the detail of the tax code so the notion of profit for the tax corresponds to the notion of profit that the company seeks to maximise.

As we’ve seen earlier in the course, there are many important counter examples. Collective action will always be required to supply “public goods” (for example defence) and to correct the distribution of income. Markets may fail altogether in some circumstances - where important information is unobtainable, for example⁴. Frequently the price that a free market would produce would not reflect the true marginal cost of things; so petrol pollutes, alcohol causes crime and illness, and so on - these are “externalities” that are not priced in in private transactions. In these cases there is a good argument for for having taxes that deliberately move the prices away from its free market level to reflect the full ‘social’ cost.

There is another sense in which we might want changes to the fiscal system to be fiscally neutral: we might want any change to have a net zero cost, so any tax cut is paid for by a spending reduction somewhere, and any benefit increase paid for by a tax rise. Proposals that fail to to this are often criticised in the media and parliament, but there might be good reasons to design a package of measures that is not zero-cost, for example if we want to boost demand in a time of recession, or if we believe your measures might improve the efficiency of the economy in the long run. In any event, it’s always important to know what the projected net cost of proposals are, and to have a good story to hand if the cost is not zero.